Sinai Peninsula is considered a very important region for Egypt at both the national and strategic levels. One of the important problems facing the researches of Sinai development there is not available detailed information about population and their distribution, so this is still a research topic. Sinai population represented approximately 0.6% from Egypt population and the total area of Sinai is 60000 square kilometer. In Egypt, Census years were 1947, 1960, 1966, 1976, 1986, 1996 and 2006. This means that every 10 years there was a new up-to-date census information and population. In between forecasting can be used to estimate population based on the available data and the used mathematical model. Census process is very hard and timely and costly consuming. Rapid growth of population in developing countries especially in Egypt leads to search alternative methods to estimate population. The main objectives of this research is using remote sensing images with different resolution used to estimate population up to date to help the decision maker to prepare the developing planning. The results of first phase of research project 2011 - 2012 is producing ortho-image (WGS84 ellipsoid and UTM projection) from digital photogrammetric workstation. The research proposed using ortho-images and digital maps produced from single and stereoscopic satellite images high resolution satellite images with different resolution and from single very high resolution satellite images for estimating population using novel and developed method to be suitable for the characteristic of residential buildings especially in Norther Sinai Governorate and generally for similar study areas. Also, an evaluation accuracy of population estimation using different novel methods has been done based mainly on predication of population for the target year to be used as reference from the available census data for each case. The results of study indicated that percentage of error in population estimation based on urban area for Markas Arish 1 has been found 8.27% using linear regression model and 7.17% using second order regression model. Percentage of error in population estimation based on urban area for Qism Markas Ber Abd has been found -18.86% using linear regression modelsِ Also, percentage of error in population estimation based on the number of housing units from very high resolution satellite images for Hasna city and Nekhel city have been found +13.7% and +9.33%. One of the outcomes is production of population density map for 2006 and 2012. |
